On Tue, 27 Jan 2015, Michal Hocko wrote:

Admittedly, I was checking only SLAB allocator when reviewing and
assuming SLUB would behave in the same way :/
But maybe I have misinterpreted the slab code as well and
get_node(struct kmem_cache *, int node) returns non-NULL for !online
nodes.

Oh. Just allocate from node 12345 in SLAB and you will also have a strange
failure.

I have briefly checked the code and it seems that many users are aware
of this and use the same construct Johannes used in the end or they use
cpu_to_node. But then there are other users doing:
net/openvswitch/flow_table.c:
        /* Initialize the default stat node. */
        stats = kmem_cache_alloc_node(flow_stats_cache,
                                      GFP_KERNEL | __GFP_ZERO, 0);

and this can blow up if Node0 is not online. I haven't checked other

Node 0 is special in many architectures and is guaranteed to exist.
PowerPC is a notable exception which causes frequent issues with NUMA
changes.

That being said I have no problem with checking node_online in the memcg
code which was reported to blow up here. I am just thinking whether it
is safe to simply blow up like that.

Node numbers must be legitimate in order to be used. Same thing with
processor numbers. We cannot always check if they are online. The numbers
in use must be sane. We have notifier subsystems that do callbacks to
allow subsystems to add and remove new nodes and processors. Those should
be used to ensure that only legitimate node and processor numbers are
used.


--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>